1. Field of the Invention
This invention relates generally to methods and apparatus for transforming thermal energy from a heat source into mechanical and then electrical form using a working fluid that is expanded and regenerated. This invention further relates to a method and apparatus for improving the thermal efficiency of a thermodynamic cycle.
2. Brief Description of the Background Art
It is well known that, in accordance with the Second Law of thermodynamics, the exergy (energy potential) of any heat source is increased as the temperature of this heat source is increased. Because of this effect, technological improvements in power generation have been directed toward increasing the temperature of the heat released in the process of combustion. One such improvement is the counterflow preheating of the combustion air with combustion gases to increase the combustion temperature and the average temperature of heat released from the burning of fuel. This technique, referred to as "pulverized-coal combustion," is well known and widely established.
Unlike the energy potential of the heat source, the efficiency of a power cycle depends, not on the temperature of the heat source directly, but on the average temperature of the working fluid in the process of heat transfer from the heat source. If this temperature of heat acquisition is significantly lower than the temperature of the available heat source, irreversible losses of exergy occur in the process of heat transfer, and the efficiency of the cycle remains relatively low.
This effect explains the relatively low efficiency of conventional power plants. For example, the limit of efficiency of a power plant converting thermal energy into power is on the level of approximately 63% , even when the working fluid temperature is maintained at the 1,000.degree. to 1,100.degree. F. limit that the metallurgical properties of modern power plants dictate. Similarly, the efficiency of the best direct-fired plants, based on a turbine electrical-power output (from which the work of the circulating feed pumps is subtracted) does not exceed 41-42%. In other words, the thermodynamic efficiency of these plants does not exceed 65% (the ratio of the thermal efficiency to the thermodynamic limit of efficiency).
The theoretical reason for this phenomenon is that the bulk of the heat transferred to the working fluid, i.e., water, is acquired in the boiler, where water boils at a temperature of approximately 660.degree. F. (350.degree. C.), while the available heat has a much higher temperature. It is absolutely clear, from a thermodynamic point of view, that unless the temperature of the heat acquisition by the working fluid is increased drastically, the efficiency of the process of conversion of thermal energy into power, i.e., the efficiency of the thermodynamic cycle, cannot be increased.
Use of a working fluid with a boiling temperature higher than that of water would not as a practical matter improve efficiency of the cycle for the following reason. The pressure in the condenser must be maintained at deep vacuum, even when water is used as a working fluid. If fluid with a normally higher-than-water boiling temperature is used, an even deeper vacuum in the condenser would be required, which would be technically impractical. Unless this super-low pressure in the condenser was provided, the temperature of condensation of such a hypothetical high-boiling fluid would be high, and all the gains obtained in the boiler would be lost in the condenser. Because of this problem, very little progress has been made in improving the efficiency of direct-fired power plants in the last sixty to seventy years.
A promising way to increase the efficiency of a power cycle utilizing high-temperature heat sources would be to use the so-called "recuperative cycle". According to this idea, the working fluid should be preheated to a relatively high temperature by the returning streams of the same working fluid. Only after such preheating should the external heat be transferred to the working fluid. As a result, all heat acquisition would occur at a high temperature, and theoretically the efficiency of such a cycle would be increased.
The only practical example of such a cycle is the so-called "recuperative Brighton Cycle", which utilizes a gaseous working fluid. In this cycle, the working fluid is compressed at ambient temperature, preheated in a recuperator, additionally heated by a heat source, expanded in a turbine, and sent back into the recuperator, thus providing preheating.
Despite its theoretical advantages, the recuperative Brighton Cycle does not, in reality, provide a superior efficiency because of two factors:
(1) the "work of compression" of a gaseous working fluid is very high and cannot be performed isothermally or with a small rise in temperature; and PA1 (2) because a gaseous working fluid is used, the temperature difference in the recuperator must be relatively high, thus causing irreversible exergy losses.
The ideal solution to a high-efficiency power cycle would be to combine a high degree of recuperation, characteristic of the Brighton Cycle, with a steam cycle wherein the working fluid pressure is increased while this fluid is in a liquid state. This allows the use of pumps, with a relatively minor work requirement (low "work of compression") to increase fluid pressure.
The direct realization of such a cycle unfortunately appears impossible, for a very simple reason. If the process of recuperative heating includes liquid preheating, evaporation, and some superheating, then the returning stream, which must have a lower pressure than the oncoming stream, would condense at a lower temperature than that at which the oncoming stream boils. This phenomenon appears to make the direct recuperation of heat in such a process impossible.
As indicated above, the overall boiling process in a thermodynamic cycle can be viewed for discussion purposes as consisting of three distinct parts: preheating, evaporation, and superheating. With conventional technology, the matching of a heat source and the working fluid is adequate only during the high temperature portion of superheating. The inventor of the present invention has appreciated, however, that in previously known processes a portion of the high temperature heat which would be suitable for high temperature superheating is used instead for evaporation and preheating. This causes very large temperature differences between the two streams, and as a result, irreversible losses of exergy. For example, in the conventional Rankine cycle, the losses arising from mismatching of the enthalpy-temperature characteristics of the heat source and the working fluid would constitute about 25% of the available exergy.
The ideal solution to the age old dilemma of poorly matched heat source and working fluid enthalpy-temperature characteristics would be one that makes high temperature heat available from the heat source for use in superheating thereby reducing the temperature differences during superheating, but at the same time provides lower temperature heat which minimizes the temperature differences in the process of evaporation.
Conventional steam-power systems provide a poor substitute for this ideal system. This is because the heat provided by the multiple withdrawal of steam, that has been partially expanded in a turbine, may only be used for the low temperature pre-heating of the incoming or feed water stream to the turbine. This use of the multiple withdrawal of steam to provide heat to the feedwater is known as feedwater preheating. Unlike its use in low temperature pre-heating, the withdrawal of partially expanded steam can not provide heat for the high temperature protion of the preheating process or for the evaporation of or for the low temperature portion of the superheating of the feedwater stream.
Because of technological limitations, the water usually boils at a pressure of approximately 2,500 psia and at a temperature of about 670.degree. F. Thus, the temperature of the heat source of these systems is generally substantially greater than the boiling temperature of the liquid working fluid. Because of the difference between the high temperature of the combustion gases and the relatively low boiling temperature of the working fluid, conventional steam systems use high-temperature heat predominantly for low-temperature purposes. Since the difference between the temperature of the available heat and the temperature required for the process is very large, very high thermodynamic losses result from an irreversible heat exchange. Such losses severely limit the efficiency of conventional steam systems.
Replacing conventional systems with a system that provides lower temperature heat for evaporation of the working fluid may substantially reduce thermodynamic losses resulting from evaporation. Reducing these losses can substantially increase the efficiency of the system.